The mutation process ultimately defines the genetic features of all populations, and hence has a bearing on the full range of issues in evolutionary genetics, inheritance, and genetic disorders. Yet, despite the centrality of mutation to biology, formidable technical barriers have constrained our understanding of the rate at which mutations arise and the molecular spectrum of their effects. The proposed research takes advantage of newly emergent technology for mutation detection to address a number of central issues. First, performing whole-genome sequencing with long-term mutation-accumulation lines, we will ascertain the rate and full molecular spectrum of spontaneously arising mutations (germline replication errors) in a wide range of eukaryotic microbial species. This work will complete a phylogenetically wide survey of mutation-rate variation (involving ~50 species), test the hypothesis that microbial eukaryotes have extraordinarily low mutation rates, and contribute to an emergent general theory on mutation-rate evolution (the drift-barrier hypothesis). Second, applying a newly developed method to the full set of study species, we will estimate the rate at which errors arise during transcription of DNA to RNA. Preliminary data suggest that such transient errors arise at rates several orders of magnitude higher than those during replication. If confirmed, this would imply that a large fraction of gene transcripts contain errors; and we will also test whether transcriptin and replication error rates scale in the same way in accordance with the drift-barrier hypothesis. Third, using the same method, we will estimate the somatic mutation rate in several tissues in a variety of multicellular species, testing the hypothesis that such rates greatly exceed those in the germline. Because mutations are the ultimate source of all inherited genetic disorders and of somatically acquired cancers, the results of this work will be of central relevance to a wide array of human-health related issues. In addition, the unbiased rates and spectra of mutation that emerge from this study will provide a powerful resource for evolutionary geneticists concerned with the sources of natural variation at the molecular level.